Universality of the limit shape of convex lattice polygonal lines (Q653304)

A convex lattice polygonal line \(\Gamma\) is a piecewise linear path on the plane, starting at the origin \(O=(0,0)\), with vertices on the integer lattice \({\mathbb Z}_+^2=\{(i, j)\in {\mathbb Z}^2:\, i, j\geq 0\}\), and such that the inclination of its consecutive edges strictly increases staying between 0 and \(\pi/2\). Let \(\Pi\) be the set of all convex lattice polygonal lines with finitely many edges, and denote by \(\Pi_n\subset \Pi\) the subset of polygonal lines \(\Gamma\in \Pi\) whose right endpoint is fixed at \(n=(n_1, n_2)\). The authors are concerned with the problem of limit shape of ``typical'' \(\Gamma\in \Pi_n\) as \(n\to \infty\) with respect to some probability measure \(P_n\) on \(\Pi_n\). The limit shape and its existence can depend on the probability law \(P_n\). With respect to the uniform distribution on \(\Pi_n\), the problem was solved independently by \textit{A. M. Vershik} [Funct. Anal. Appl. 28, No. 1, 13--20 (1994); translation from Funkts. Anal. Prilozh. 28, No. 1, 16--25 (1994; Zbl 0848.52004)], \textit{I. Bárány} [Discrete Comput. Geom. 13, No. 3--4, 279--295 (1995; Zbl 0824.52001)] and \textit{Ya. G. Sinai} [Funct. Anal. Appl. 28, No. 2, 108--113 (1994); translation from Funkts. Anal. Prilozh. 28, No. 2, 41--48 (1994; Zbl 0832.60099)]. The proofs in the papers of Vershik [loc. cit] and Bárány [loc. cit.] involved a blend of combinatorial, variational and geometric arguments and were based on a direct analysis of the corresponding generating function via a multivariate saddle-point method for a Cauchy integral (see [Vershik, loc. cit.]) or a suitable Tauberian theorem (see [Bárány, loc. cit.]). Extending some of these ideas and using large deviations techniques, \textit{A. Vershik} and \textit{O. Zeitouni} [Isr. J. Math. 109, 13--27 (1999; Zbl 0945.60022)] developed a systematic approach to the limit shape problem for the uniform measure on more general ensembles of convex lattice polygonal lines with various geometric restrictions. Sinai [loc. cit.] proposed an alternative, probabilistic method essentially based on randomization of the right endpoint of the polygonal line \(\Gamma\in \Pi_n\). Vershik [loc. cit, p. 20], pointed out that it would be interesting to study asymptotic properties of convex lattice polygonal lines under other probability measures \(P_n\) on \(\Pi_n\), and conjectured that the limit shape might be universal for some classes of measures. The authors of the present paper prove Vershik-Prokhorov's universality conjecture for a parametric family of probability measures \(P_n^r\), even though \(P_n^r\), \(r\neq1\) and \(P_n^1\) are asymptotically singular. Measures \(P_n^r\) are constructed following Sinai's approach. The proofs involve subtle analytical tools including the Möbius inversion formula and properties of zeros of the Riemann zeta function.